Evaluate the impact of mass media on breastfeeding practices in vietnam

Evaluation plan and baseline indingsDecember 2012 Using behavioral theory to evaluate the impact of mass media on breastfeeding practices in Viet Nam... Using behavioral theory to eva

Evaluation plan and baseline indings

December 2012

Using behavioral theory

to evaluate the impact of mass media on breastfeeding

practices in Viet Nam

Alive & Thrive (A&T) is a six-year (2009-2014) initiative to improve infant and young child feeding practices by increasing rates of exclusive breastfeeding and improving complementary feeding practices The irst 2 years of life provide a window of opportunity to prevent child deaths and ensure healthy growth and brain development Alive & Thrive aims to reach more than 16 million children under 2 years old in Bangladesh, Ethiopia, and Viet Nam through various

delivery models Learnings are shared widely to inform policies and programs throughout the world Alive & Thrive is funded by the Bill & Melinda Gates

Foundation and managed by FHI 360 Other members of the A&T consortium include BRAC, GMMB, IFPRI, Save the Children, World Vision, and UC-Davis.

Recommended citation

Alayón, S., Naugle, D., Jimerson, A., Lamarre-Vincent, J., Baume, C (2012) Using behavioral theory to

evaluate the impact of mass media on breastfeeding practices in Viet Nam: Evaluation plan and baseline indings Washington, DC: Alive & Thrive.

Alive & Thrive

FHI 360

A C K N O W L E D G M E N T S

Many contributed to the concept for this

evaluation and to the design of the evaluation

plan for Alive & Thrive’s Viet Nam mass media

campaign for improved breastfeeding practices

In particular, we thank Dr Carol Baume,

inde-pendent consultant and expert in mass media

for health evaluation, for leading the way in

developing the evaluation plan Dr Robert Hornik

of the Annenberg School for Communication,

University of Pennsylvania, generously

con-tributed his time to review our draft plan and for

a provocative discussion that helped us reine the

design Dr Hornik also steered us to Danielle

Naugle, a doctoral student in his program, who

added to the design and initiated the analyses

for this report

We acknowledge the contributions of IFPRI,

Alive & Thrive’s partner in measurement,

learning, and evaluation, especially the work of

Dr Phuong Hong Nguyen, to reinement of the

baseline questionnaire Others who helped shape

the questionnaire are: staff of the Institute of

Social and Medical Studies (ISMS) in Viet Nam,

under the direction of Dr Nam Nguyen; Nemat

Hajeebhoy, A&T Viet Nam senior country

director and her staff, including Thi Mien Nguyen;

and the team of Carol Baume, Danielle Naugle,

and Ann Jimerson

ISMS and a team of supervisors and data

collectors ably conducted that baseline survey

and provided initial data runs

Early analyses of the baseline data were

conducted by Dr Zo Rambeloson and Dr Tuan

Thanh Nguyen, both of FHI 360 Danielle

Naugle, with support from University of Pennsylvania, conducted later analyses, especially focusing on behaviors and their determinants

Dr Kelly L’Engle and Mr Jesse Lamarre-Vincent

of FHI 360’s Social and Behavioral Health Sciences (SBHS) department offered valuable insights, further analyses, and written summaries

of the indings

Many FHI 360 staff reviewed drafts of this report, including: Jean Baker, Nemat Hajeebhoy, Giang Nguyen Huong, Jesse Lamarre-Vincent, Kelly L’Engle, Nadra Franklin, Luann Martin, Danielle Naugle, Tuan Thanh Nguyen, and Kim Winnard

We thank Nadra Franklin of FHI 360 for herongoing leadership to ensure the rigor thisevaluation requires Jill Vitick's design makes the report appealing and easy to understand

We are grateful to the Ministry of Health, the Department of Maternal and Child Health, and the National Institute of Nutrition in Viet Nam for enabling us to undertake this research at the selected sites

We also want to acknowledge the Bill & Melinda

Gates Foundation for providing additional fund-

ing to Alive & Thrive for the Viet Nam mass

media campaign and evaluation

And, of course, this evaluation would not be

possible without several thousand Vietnamese

mothers who trusted us enough to share details

about their lives and their children

Silvia Alayón and Ann Jimerson Alive & Thrive

Washington, DC, 2012

A C R O N Y M S

BF Breastfeeding

T A B L E O F C O N T E N T S

Acknowledgments i

Acronyms ii

Executive Summary iv

Background 1

Evaluation Design .1

Viet Nam Context 3

Evaluation Plan 8

Baseline Survey 8

Baseline Results 10

Sample Description 10

Exclusive Breastfeeding 11

Behavioral Determinants 13

Applications 21

Bibliography .23

L I S T O F T A B L E S Table 1: Sample characteristics 10

Table 2: Exposure to media and information .11

Table 3: Mothers practicing exclusive breastfeeding and its component behaviors 12

Table 4: Exclusive breastfeeding by background characteristics 17

Table 5: Exclusive breastfeeding by exposure to other sources of information 18

Table 6: Mapping determinants of exclusive breastfeeding to the mass media messages 21

L I S T O F F I G U R E S Figure 1: A&T’s behavior change model for mass media campaign 4

Figure 2: Media burst timeline 7

Figure 3: Exclusive breastfeeding and component behaviors, by age of child (months) 12

Figure 4: Mothers’ knowledge about breastfeeding 14

Figure 5: Mothers’ beliefs about behavioral control 15

Figure 6: Mothers’ beliefs about the consequences of select behaviors 16

ANNEX A: Evaluation approaches 26

ANNEX B: Scripts from Alive & Thrive’s two TV spots on breastfeeding, Viet Nam mass media campaign .28

ANNEX C: Associations between behavioral determinants and behaviors: Odds ratios 29

E X E C U T I V E S U M M A R Y

The Bill & Melinda Gates Foundation asked

the question: To what extent can a national

mass media campaign alone change

breast-feeding behaviors?

Alive & Thrive Viet Nam had, from the initial

program design, planned to take advantage of

the country’s sophisticated media environment in

its promotion of improved breastfeeding

prac-tices In a country of more than 87 million, TV

viewership is almost universal and Internet use

is growing by leaps and bounds Mass media is a

logical tool for reaching enough families to have

a population-wide effect on health behaviors Yet

there is a dearth of evidence on mass media’s

impact on breastfeeding A 2010 Lancet review of

mass media for health promotion noted that, for

breastfeeding media campaigns, “reviews from

the 1990s onwards seem scarce or non-existent.”

(Wakeield, Loken, & Hornik, 2010)

The Foundation was eager to invest in a mass

media campaign, but asked that a portion of

the investment be used to mount a rigorous

evaluation of that campaign to shed light on the

role of mass media in effecting behavior change

on a large scale This report lays out that

evalua-tion plan and shows how a series of surveys over

3 years will assess the media campaign’s impact

Initial analyses of the baseline data show how

these early indings are aiding in reining this

mass media evaluation plan and in identifying

the precise triggers of breastfeeding behaviors

In this quickly industrializing country, only

about 20 percent of infants under 6 months are

exclusively breastfed, as recommended by WHO

Exclusive breastfeeding is not a behavior in itself,

but the result of a series of component behaviors:

not offering water or other liquids, not giving

infant formula, and not feeding solid or semisolid

foods The evaluation will capture changes in

Numerous factors inluence a mother’s feeding practices Alive & Thrive’s (A&T’s) formative research teased out the complex factors that tip mothers toward giving up too early on breastfeeding In Viet Nam, the baby’s grand-mothers and father hold powerful sway Health providers may tell the mother she is incapable

breast-of producing enough milk, even for baby’s irst day The mother watches what others like her are doing Despite restrictions on advertising of breastmilk substitutes, she is bombarded with marketing of infant formula

Alive & Thrive’s mass media evaluation is built

on behavioral theory and formative indings

It includes creation, testing, and reinement

of questions to measure the likely behavioral determinants The analyses of baseline data demonstrate that many of the speciic factors measured, to assess knowledge and the mother’s beliefs towards the behaviors—her attitudes, perceived social norms, and perceived behavioral control, are strongly associated with exclusive breastfeeding and its component behaviors Yet

in at least one case, quantitative data from the baseline survey show that a particular factor assumed to prompt behavior change is not, after all, associated with the behaviors The evaluation will provide ongoing tracking of mothers’ knowl-edge and beliefs Sharing of the methods reined for measuring these determinants and the early indings can help move the nutrition ield forward into practical application of behavioral science to strategic program design

A&T’s mass media campaign beneits from high level expertise, starting with support from the cross-cultural marketing irm Huemanitas to gain insights into Vietnamese cultural constraints and supports for breastfeeding Market research and pretesting by Nielsen’s Viet Nam afiliate, media development and production by Ogilvy &

Mather, and media placement by MAXUS Viet

live, talking babies doling out advice to their

mothers on how and why to breastfeed

exclu-sively A popular and beautiful actress plays the

loving mother, and colors, visuals, and the sweet

voices of children make the spots emotionally

appealing The irst spot addresses the need to

avoid giving babies water and afirms that with

breastmilk alone, the baby will get enough liquid

to avoid being thirsty or hot The second spot

assures mothers that they can produce breastmilk

that is high enough in both quality and quantity

to offer their babies complete nourishment for 6

months Both spots stress messages that testing

showed appeal to a Vietnamese mother:

breast-milk contains all the water and nutrients babies

need for 6 months; exclusive breastfeeding makes

children smart and healthy; the advice is based

on scientiic evidence from global experts; and

mothers like her trust in exclusive breastfeeding

The baseline survey, which included 6,175

mothers of infants 0-5.9 months in 11 provinces,

measured mothers’ reported exposure to different

sources of information on and support for

breast-feeding Interim surveys will measure individual

mothers’ exposure to the media campaign’s

messages and allow for checking associations

between exposure and behaviors as well as

between exposure and behavioral determinants

Scholars offer evidence that public health

com-munication—even mass media alone—can result

in positive changes in behaviors (Hornik, 2002)

With insights from leaders in the ield of munication evaluation, including experts

com-at Annenberg School for Communiccom-ation/

University of Pennsylvania, A&T developed

a solid evaluation plan that accommodates the reality of an evolving mass media campaign

Since mass media reach is almost universal, there is no way to establish a control group that

is not exposed to the campaign’s messages

To compensate for lack of a control group, the evaluation plan comprises four approaches Each approach has methodological limitations But if the approaches support consistent conclusions, together they can provide strong evidence for campaign effects This report describes those four approaches

Baseline indings alone, of course, cannot answer the question about what effects a media campaign can have on breastfeeding behaviors

But the analysis of this baseline data does begin

to clarify the design of the mass media evaluation plan itself Already, data analysis has allowed for reinement and reduction of the questions used to measure behavioral determinants, and preliminary establishment of associations between those determinants and the behaviors promoted This report on the evaluation plan and baseline indings offers support for others as they consider ways to use and evaluate mass media for improved infant and young child feeding

Using behavioral theory to evaluate the impact

of mass media on breastfeeding practices in

Viet Nam: Evaluation plan and baseline indings

By Silvia Alayón, Danielle Naugle, Ann Jimerson, Jesse Lamarre-Vincent, and Carol Baume

B A C K G R O U N D

By 2008, global health experts had the science

to conirm that proper nutrition can prevent one

in ive deaths of children under ive years of age

If mothers were to practice early and exclusive

breastfeeding for the baby’s irst 6 months of

life, it could afford infants a six times greater

chance for survival Adequate nutrition in the irst

2 years of life would save millions of lives and

set millions of children on the course to optimal

growth and development (Bhutta, et al., 2008)

Given relatively low funding levels and lack of

clarity about “what works,” most child feeding

interventions have been small, local, and limited in

scope In 2009, the Bill & Melinda Gates

Founda-tion awarded the Alive & Thrive (A&T) project to

develop models for improving infant and young

child feeding practices at scale and to document

what does and does not work to improve child

feeding, including exclusive breastfeeding.1 Three

diverse countries with large

populations—Bangla-desh, Ethiopia, and Viet Nam—were selected to

develop and implement comprehensive program

models that work through policy advocacy, mass

media, community mobilization, interpersonal

communication, and service delivery

Other health programs (e.g., HIV, family

plan-ning, and vaccine promotion) have demonstrated

the impact that traditional mass

media—televi-sion, radio, and outdoor advertising—can have

on health behaviors However, a recent review in

The Lancet of the use of mass media campaigns

to change health behavior noted: “Although

mass media programmes to promote ing have been mounted, reviews from the 1990s onwards seem scarce or non-existent” (Wakeield, Loken, & Hornik, 2010)

breastfeed-Viet Nam is a sophisticated media market with nearly universal access to television Data from

a survey conducted by A&T in four provinces in

2011 revealed that about 99 percent of mothers with children under two had access to television

Over 90 percent of them reported watching TV weekly (Nguyen P, 2011) Though A&T has invested in policy advocacy and interpersonal communication in Viet Nam, the wide reach of mass media provides a unique opportunity to deliver child feeding messages at scale In Viet Nam, A&T aims to reach nearly 1 million chil-dren under 2 years of age with intensive activities such as one-to-one and group counseling and another 1.5 million through a carefully designed mass media campaign, which aims to increase the rates of exclusive breastfeeding among infants under 6 months of age

E V A L U A T I O N D E S I G N Given the dearth of rigorous studies on the effect

of mass media on breastfeeding promotion, A&T paired the mass media campaign with a rigorous evaluation The evaluation is expected to add

to the body of knowledge on the use of mass media as an intervention to increase exclusive breastfeeding and related behaviors This report describes the approaches that are being used to evaluate the A&T mass media campaign in Viet

The goal of the evaluation is to document the

effect of the mass media campaign on

breastfeed-ing behaviors in Viet Nam A major challenge

in designing this evaluation was the nationwide

reach of the mass media campaign and the

near-universal coverage of television, which made it

impossible to establish an unexposed control or

comparison group To overcome this challenge,

A&T consulted with experts in public health

communication research2 to design an evaluation

that would allow the project to understand the

effect of its mass media campaign on exclusive

breastfeeding rates The evaluation aims to

respond to the following questions:

• To what extent can mass media alone change

breastfeeding behaviors?

• What is the impact of mass media combined

with interpersonal interventions on the

same behaviors?

The data will also allow A&T to explore

the following:

• Reach How wide is the reach of A&T’s

campaign? What percentage of the target

audience recalls seeing or hearing the

campaign messages?

• Behavior change. Can we link exposure to

the campaign to increases in speciic

breast-feeding behaviors?

• Behavioral determinants of exclusive

breastfeeding. The evaluation will contribute

to knowledge about what the important

determinants for exclusive breastfeeding are:

knowledge, attitudes, perceived advantages/

disadvantages, outcome expectation,

self-eficacy, and norms

• Role of media campaign. How does exposure

to the media campaign compare with exposure

2 Dr Robert Hornik, Wilbur Schramm Professor of Communication

at Annenberg School for Communication, University of Pennsylvania;

Dr Carol Baume, independent consultant and expert on

communi-cation evaluation; and Danielle Naugle, PhD student at Annenberg

School for Communication, University of Pennsylvania

to other elements of the program (e.g., social franchises for face-to-face nutrition counsel-ing)? How does the mass media component intersect with other components?

• Rural-urban. Does the impact of mass media vary for urban and rural populations?

A&T works in 15 of the 63 provinces and palities in Viet Nam The overall structure of the evaluation is a modiied continuous measurement design To economize, the evaluation’s baseline data collection in 2011 was coordinated with a previously planned survey in 11 provinces Three interim surveys, in 2012 and 2013, are timed to correspond to the media bursts An endline survey will be conducted in 2014

munici-To compensate for the inability to establish a control group, four different approaches will be integrated into the mass media evaluation Each

of the approaches, briely described below, has methodological limitations However, if the approaches support consistent conclusions, together they can provide strong evidence for campaign effects The irst approach simply assesses the rate of change in the outcome of interest—exclusive breastfeeding—before, during, and after the campaign Data are avail-able from before the campaign, and A&T will have several data points during the campaign

to construct these trends The expectation with this approach is that the slope of the trend in breastfeeding rates observed during and after the campaign will be steeper than the slope observed before the media campaign The second approach will explore whether there is an association between exposure to the campaign and exclusive breastfeeding, while controlling for as many confounders as possible This approach also measures whether knowledge, beliefs, and social norms addressed by the campaign are associated with exposure and behavior, more so than those that are not addressed by the campaign The

third approach is a constructed cohort design;

it will use baseline data to predict exposure to

the mass media campaign Based on predicted

exposure, individuals will be classiied into two

groups—high and low exposure—and their rates

of exclusive breastfeeding will be compared

over time In a successful campaign, the rate of

exclusive breastfeeding is expected to increase

more quickly in the high exposure group The

fourth approach is also a constructed cohort

design, using baseline data to create and compare

high and low exposure geographical areas All of

these approaches are described in more detail in

Annex A

V I E T N A M C O N T E X T

Although Viet Nam is a lower middle-income

country, stunting of children under 5 years of

age is still prevalent In 2010, the Viet Nam

National Institute of Nutrition (NIN) reported that

29 percent of children under 5 years of age are

stunted While most mothers (98 percent) in Viet

Nam breastfeed for a period of time, practices are

suboptimal Many mothers give their children

water, use infant formula, and/or introduce soft

and semisolid foods too early (i.e., before 6

months of age) Only 62 percent practice “early

initiation,” putting the baby to breast within

the irst hour of life Even fewer, 20 percent of

infants under 6 months of age, are exclusively

breastfed, as recommended by WHO and

UNICEF (National Institute of Nutrition,

UNICEF, Alive & Thrive Viet Nam, 2010)

Alive & Thrive Viet Nam

In Viet Nam, A&T has established a network

of almost 800 Mat Troi Be Tho (MTBT) or The

Little Sun franchises to deliver individual and

group counseling on infant and young child

feeding (IYCF) within existing health centers

In addition, IYCF support groups have been

established in 225 villages not served by the franchises All interpersonal contacts are designed for pregnant women and mothers and other caretakers of children under 24 months

Even with this extensive network to deliver interpersonal communication interventions at scale, A&T estimated that an additional 1.5 mil-lion mothers would need to be reached to achieve population level changes in exclusive breastfeed-ing rates To reach the intended target, A&T took advantage of high rates of TV viewership and designed and aired a national campaign that promotes exclusive breastfeeding for 6 months

From A&T’s initial planning stages in Viet Nam, mass media advertising through various channels was viewed as a valuable tool for large-scale promotion of improved feeding practices It was assumed that investing heavily in mass media, along with strengthening regulation and compli-ance with the Code of Marketing of Breastmilk Substitutes could shape breastfeeding practices

in a rapidly urbanizing environment where infant formula was heavily advertised and where women’s work outside the home was perceived to

be a major barrier to exclusive breastfeeding

A&T Mass media strategy in Viet Nam

The campaign’s strategy was based on a ied behavior change model, which represents

simpli-a synthesis of behsimpli-avior chsimpli-ange theories (see Figure 1 on the next page) The model proposes that exposure to the campaign’s messages leads

to a change in behavioral determinants such as knowledge, beliefs about outcomes of the behav-ior, perceptions of social norms, and self-eficacy

These, in turn, inluence breastfeeding behaviors

The model suggests that exclusive breastfeeding

is not a single behavior, but a collection of behaviors, referred to hereafter as “component behaviors.” Each woman’s decision to practice the component behaviors is inluenced by the

Behavioral Determinants

Exposure to Messages

on Breastfeeding

Knowledge

Beliefs about consequences

Beliefs about social norms

Beliefs about behavioral control

Behaviors

Do not give water

Do not give infant formula

Do not give other liquids

(news stories placed)

A&T media campaign

TV spots

Breastfeed exclusively for 6 months

Do not give semisolid

or solid foods Component behaviorsFIGURE 1: A&T’S BEHAVIOR CHANGE MODEL FOR MASS MEDIA CAMPAIGN

knowledge she possesses, her beliefs about the

potential beneits of practicing the behavior,

her beliefs about social norms and the value

she places on these when making decisions,

and her beliefs about her own ability to practice

these behaviors These factors are referred to as

“behavioral determinants” and are believed to be

precursors to behavior change The mass media

messages in Viet Nam aim to increase exclusive

breastfeeding by both promoting the component

behaviors and also addressing the behavioral

determinants believed to be most strongly

associ-ated with exclusive breastfeeding

A&T formative research informs design of

the mass media strategy

In 2009, A&T conducted formative research

which helped program staff identify patterns

of exclusive breastfeeding and its component

behaviors (Hoat, Huong, and Xuan, 2010) Both

qualitative and quantitative studies identiied

current breastfeeding practices and the barriers and facilitators of optimal practices The study conirmed that exclusive breastfeeding falls off dramatically over time By 5 months of age, very few children are exclusively breastfed

In Viet Nam, giving infants water after breastfeeding is a common practice, either to quench thirst or to “clean” the infant’s mouth after feeding In the formative research, mothers often stated that they give water because they think the baby will be thirsty, especially in hot weather They also believe they must rinse their baby’s mouth out to avoid oral thrush, a yeast infection in the mouth common among newborns

Giving water was identiied as the main barrier to exclusive breastfeeding Another major challenge

to 6 months of exclusive breastfeeding identiied

in the study was, for those employed in the formal sector (20 percent-30 percent), the need to return

to work after 4 months of maternity leave

Findings showed that only half of mothers

understood what “exclusive breastfeeding”

meant (that is, only breastmilk and nothing else),

meaning that media messages would need to be

explicit about the component behaviors, such as

not giving water Among mothers and those who

support and guide them, lack of conidence (or

self-eficacy) that their breastmilk is suficient

leads many women to add infant formula as a

hedge against their “poor quality” breastmilk

(Hoat, Huong, & T, 2010) Many mothers said

they believed they lacked suficient breastmilk

(in both quantity and quality) to nourish their

children adequately up to 6 months of age,

contributing to use of formula and to early duction of complementary foods Mothers stated that even health providers doubted the mothers’

intro-ability to produce adequate breastmilk Interviews with health workers revealed that this doubt led them occasionally to advise mothers to take infant formula to the delivery site or to give the baby infant formula beginning around 3 or 4 months

of age Mothers revealed that social norms play

an important role in their breastfeeding practices, stating that their breastfeeding decisions were inluenced by other mothers, parents, other family members, and advertisements

The internationally-established definition of exclusive breastfeeding

is feeding an infant breastmilk alone, with no other water, liquids,

or foods Therefore exclusive breastfeeding is a set of “component

behaviors,” each of which needs to be practiced if the exclusive

breastfeeding goals are to be achieved The component behaviors

are described below

• Not giving water Vietnamese mothers and caretakers give babies water, both as

a drink to quench baby’s thirst and as a rinse to clean baby’s mouth after feeding

because they believe that not rinsing causes thrush Both uses of water put the baby

at risk of infection (from unclean water) Giving water as a drink displaces

breast-milk, thus robbing the child of needed nutrients and possibly leading to lower milk

production if the baby suckles less

• Not giving infant formula Infant formula displaces breastmilk, may lessen

breastmilk production, and exposes baby to infection when unclean water is used

to prepare formula

• Not giving other liquids Other liquids displace breastmilk.

• Not giving semi-solid or solid foods Introducing foods too early will displace

breastmilk and expose baby to infection

The second phase of the formative study (Alive

& Thrive, 2012b) (TIPs, or trials of improved

practices) identiied the following barriers to

exclusive breastfeeding for mothers with babies 2

or 3 months old: family and community pressure,

lack of appropriate information, dificulty in

changing ingrained practices, and lack of time

and convenience Mothers of children 4 to 5

months old identiied similar barriers to exclusive

breastfeeding and also listed mothers’ return to

work and the perception that breastmilk is not

suficient for children at this age Mothers

indi-cated that factors that could facilitate improved

breastfeeding practices included receiving clear

explanations as to why infants need exclusive

breastfeeding and why they do not need water

To begin developing a strategic plan for the

mass media campaign, A&T partnered with a

U.S.-based cross-cultural marketing irm,

Hue-manitas, to review the formative research indings

and to supplement that data with a cultural

immersion exercise to capture Vietnamese

com-munication styles, family dynamics, aspirations,

cultural narratives, and media engagement The

indings from Huemanitas’ study and

consulta-tions led to testing of a series of concepts or

message platforms They showed A&T that

the media campaign should use emotionally

appealing messages and emphasize that:

• Exclusive breastfeeding is associated with the

child’s intelligence (a highly revered beneit)

• Global, scientiic experts recommend

exclusive breastfeeding for 6 months

(a speciic social norm)

• Other Vietnamese mothers breastfeed

exclusively (social norm)

• Mothers can produce milk of high enough

quality and quantity to provide all the nutrients

and water the baby needs for 6 months

(behavioral control)

A&T took these initial concepts to Ogilvy &

Mather Viet Nam for them to design a munication strategy and two initial TV spots with accompanying print, bus advertisements, and neighborhood loudspeaker announcements An extensive period of concept testing and pretesting

com-of messages and materials, conducted by Nielsen, led to selection of an attention-getting “live talking babies” approach

The irst two TV spots, in 30-second and 45-second versions, focus on the importance

of not giving baby water in the irst 6 months (Spot #1, “No Water”) and on building mothers’

conidence that, with frequent feedings, they can produce adequate breastmilk (Spot #2,

“Nurse More”) The TV campaign was launched

in November 2011 The irst two TV spots on breastfeeding were shown on two national and

15 provincial TV channels and were broadcast throughout 2012, in three bursts See Figure 2 on the next page for timing of media bursts

In addition to TV broadcasts of the spots, the

mass media campaign includes digital and

out-of-home advertising, such as bus wrap advertising,

mobile technologies, and village loudspeaker

systems to expose target audiences to key

mes-sages The TV spots are promoted through the

Internet on 24 popular websites in Viet Nam

In August 2011, before the TV campaign was

launched, A&T and the National Institute of

Nutrition (NIN) launched the Mat Troi Be Tho

website, a resource featuring educational short

ilms and audio clips about breastfeeding (as well

as complementary feeding) and featuring the

spots The website attracted almost 31,500 visits

from August 19 to December 1 of 2011

Through the A&T media fellowship and award

program, 96 articles on IYCF were placed in

print and electronic newspapers, and 11 TV news stories were broadcast on national and local channels over 6 months in 2011

Spot 1 = No water Spot 2 = Nurse more

S = LCD/TV Screening in supermarkets and hospitals = Proposed media burst

I = Internet

Spots:

1, 2 Spots:

Screening in supermarkets and hospitals

FIGURE 2: MEDIA BURST TIMELINE

E V A L U A T I O N P L A N

The plan for the evaluation of the mass media

campaign calls for two intensive measurement

waves—the baseline conducted in July 2011 and

endline planned for 2014 in 11 A&T provinces In

addition, in October 2012 and April and

Septem-ber 2013, three interim surveys will be conducted

in a subset (4) of the 11 baseline and endline

provinces Each survey will collect information

about exposure to the campaign (except for the

baseline) and other sources of infant feeding

information, behavioral determinants, and

breastfeeding behaviors Each of these surveys

will be conducted in areas where the only A&T

intervention is the mass media campaign and also

in areas where A&T also operates a Mat Troi Be

Tho franchise.

The baseline survey occurred just before the

launch of the TV campaign, and the interim

the TV spots The endline survey will occur as soon as possible after the TV spots go off the air See igure 2 on previous page for timing and content of the media campaign’s activities and the timing of the surveys for this evaluation

B A S E L I N E S U R V E YAll of Viet Nam’s 63 provinces are reached by the mass media campaign In 15 of those, A&T

operates the Mat Troi Be Tho (MTBT) social

franchises for nutrition counseling In 2011, a survey collected data from mothers of infants and young children 0-24 months in 11 of the interven-tion provinces to provide planning data to provin-cial level governments, while also establishing the baseline for the mass media evaluation The sample included women from A&T districts as well as non-A&T districts

The women in the survey were selected using a systematic random sampling method Mothers/

children for each primary sampling unit (PSU) were selected to achieve the proposed sample size of children under 6 months for that province;

a total of 90 PSUs were included for sampling purposes The selection procedure listed children

in order by date of birth and then selected the index child using a sampling interval number, ensuring an even distribution across the child’s age The sampling methodology is described in more detail in the report for the 11 provinces (Alive & Thrive, 2012a)

The sample size for the 11 province survey was established to be large enough to detect

an increase of at least 8 percentage points in exclusive breastfeeding among mothers with children under 6 months of age in each province

For this report, the data provided by mothers of infants 0-5.9 months in 2011 was extracted from the 2011 data set and reanalyzed The sub sample examined in this report is 6,175 pairs of mothers and their children under the age of 6 months, in

The more I breastfeed, the more milk my mommy will make for me!

Mommy, I only need breastmilk

for the first six months

Mother’s milk – The best milk for us, proven globally

For mothers of children under 6 months, the

questionnaire included the following 12 modules:

background and demographic characteristics;

breastfeeding practices; complementary feeding

practices; feeding during illness; behavioral

determinants of IYCF practices; utilization of

services; media exposure; awareness, trial, and

adoption of IYCF practices; hygiene (washing

hands); household economic status; food security;

and anthropometric measurements of weight

and height

Although the ultimate goal of the mass media

campaign in Viet Nam is to increase rates of

exclusive breastfeeding, A&T recognizes that

even if the campaign fails to change breastfeeding

rates by the time of the post-survey, it may have

an effect on some of the predisposing conditions

of exclusive breastfeeding described in the

behav-ioral model Thus, each element in the model

(Figure 1) was measured This will allow A&T to

describe not just whether changes in

breastfeed-ing occurred, but also the mechanisms by which

the media may have inluenced behavior

At the time of the baseline survey, the TV

spots had not been inalized, and it was not

known which of the behavioral determinants of

breastfeeding and its component behaviors would

be used as media messages To address this,

A&T developed, tested, and incorporated into the

baseline survey a set of new measures of beliefs

that were considered likely to be determinants

of the breastfeeding behaviors that the A&T

campaign would promote Prior surveys had

measured knowledge, but not other theory-based

factors like beliefs about consequences, social norms, or behavioral control

In the sections that follow, the report will lish baseline levels of key indicators related to the mass media campaign This report explores the component behaviors of exclusive breastfeeding

estab-It aims to identify which of the determinants measured in the baseline survey are associated with exclusive breastfeeding and the strength

of the association Identifying these behavioral determinants early in the evaluation has helped

to determine which ones should continue to be measured in subsequent rounds of data collection

Mommy, I only need breastmilk for the first six months

Mother’s milk – The best milk for us, proven globally

I don't need to drink any water – my mommy's milk gives me all the water I need!

Baseline Results

S A M P L E D E S C R I P T I O N

Women in the study sample were predominantly

from the Kinh ethnic group (92 percent), married

(97 percent), and not currently working (92

per-cent) (Table 1) Approximately half of the children

0-5.9 months old were male (52 percent) and the

other half female (48 percent) In the large majority

of households of study participants, the father of

the index child was present (89 percent) as well as

the grandmother of the index child (64 percent)

Women in the study sample ranged from 16 to

47 years of age Two-thirds of participants (66 percent) were 20 to 30 years of age, with an additional quarter (24 percent) of participants between 30 and 40 years of age (Table 1) Most participants (66 percent) had completed high school, while 16 percent completed more than high school and 18 percent completed less than high school The majority of women (84 percent) lived in rural areas By design, the age in months

of the index child was equally distributed across

0 to 6 months, with between 16 percent and 19 percent of the sample in each one-month period

Sample Characteristics n = 6,175 Percent

More than high school 16%

89%

Presence in household of index child's grandmother

64%

TABLE 1: SAMPLE CHARACTERISTICS

Exposure to mass media and to information

about infant feeding

Respondents were asked about their exposure to

mass media and the frequency of this exposure

Several questions were developed to characterize

individual mothers’ current media habits in order

to proile different types of viewers Exposure to

advertising for infant formula was measured The

baseline also included a “foil” question, asking

mothers whether they had seen an image that had

never been aired in a TV spot on breastfeeding

This would help assess the degree of accuracy

of responses to be expected to future questions

about speciic TV spots Exposure to all potential

sources of information and support for exclusive

breastfeeding was measured, including advice

received from health workers

As shown in Table 2, roughly 98 percent of

respondents had ever watched TV, and about 94

percent reported watching TV at least weekly

Even before the launch of the A&T mass media

campaign, many women (39 percent) reported

having seen a message on TV about breastfeeding

in the 30 days prior to the survey Twice as many

(80 percent) reported seeing an ad for infant

formula on television during the same period

Only 7 percent reported having attended a

nutrition counseling session at a MTBT franchise

or an IYCF support group, and 6 percent reported

attending a meeting or workshop on

breastfeed-ing However, 53 percent reported that they were

shown how to breastfeed in the irst 3 days after

the birth of the index child About one third of

the mothers reported that in the 3 months prior

to the survey they had received infant feeding

advice from a nurse or doctor; and 19 percent

had received such advice from a village health

worker Certainly, even before the A&T

cam-paign, women were exposed to messages about

feeding their infants About 40 percent reported

having received feeding advice from either a MTBT franchise, a support group, or a health provider (doctor, nurse, or village doctor) in the

3 months prior to the survey (data not shown)

TABLE 2: EXPOSURE TO MEDIA AND INFORMATION

Shown how to breastfeed by someone

in the irst three days after birth 53%

Had a doctor/nurse in health facility give you advice on feeding in past

TABLE 3: MOTHERS PRACTICING

EXCLUSIVE BREASTFEEDING AND ITS

COMPONENT BEHAVIORS

Component behaviors n = 6,175

Percent of mothers who reported practicing the behavior

Breastfed exclusively, but rinsed

Did not give infant formula 69%

Did not give other liquids 79%

Did not give semi solid or solid foods 73%

WHO recommends calculating exclusive breastfeeding among children 0-5.9 months of age based on what was fed to the infant in the 24 hours prior to the survey Infants who were not fed anything other than breastmilk the previous day are considered to have been exclusively breastfed Those who were given other liquids such as water, juice, milk, and infant formula or any soft or semisolid foods are not classiied as being exclusively breastfed In Viet Nam, it is common for women to rinse the baby’s mouth with water after breastfeeding By the strictest standard, this practice would disqualify a child from being considered exclusively breastfed

To be consistent with the way that the National Institute of Nutrition has chosen to address the

Did not give complementary foods Did not give infant formula Did not give water

Did not give other liquids Practiced exclusive breastfeeding

FIGURE 3: EXCLUSIVE BREASTFEEDING AND COMPONENT BEHAVIORS, BY AGE OF CHILD (MONTHS)